Decades of behavioral endocrinology research have shown that hormones and behavior have a bidirectional relationship; hormones both influence and respond to social behavior. In contrast, hormones are often thought to have a unidirectional relationship with ornaments. Hormones influence ornament development, but little empirical work has tested how ornaments influence hormones throughout life. Here, we experimentally alter a visual signal of fighting ability in Polistes dominulus paper wasps and measure the behavioral and hormonal consequences of signal alteration in signalers and receivers. We find wasps that signal inaccurately high fighting ability receive more aggression than controls and receiving aggression reduces juvenile hormone (JH) titers. As a result, immediately after contests, inaccurate signalers have lower JH titers than controls. Ornaments also directly influence rival JH titers. Three hours after contests, wasps who interacted with rivals signaling high fighting ability have higher JH titers than wasps who interacted with rivals signaling low fighting ability. Therefore, ornaments influence hormone titers of both signalers and receivers. We demonstrate that relationships between hormones and ornaments are flexible and bidirectional rather than static and unidirectional. Dynamic relationships among ornaments, behavior, and physiology may be an important, but overlooked factor in the evolution of honest communication.honesty | challenge hypothesis | androgens | communication | social costs T he prevailing view in communication research is that hormones, such as androgens, influence the expression of ornaments and also function as "physiological costs" that maintain signal accuracy over evolutionary time (1). For example, the immunocompetence handicap hypothesis proposes that high androgen titers are required for animals to develop elaborate ornaments and high androgen titers also impose immune-related costs (2, 3). As a result, only the best individuals can afford to produce elaborate ornaments.There is growing evidence that the unidirectional view of hormone/ornament relationships is incomplete (4, 5). Instead, relationships between hormones and ornaments may be bidirectional and therefore more complex than previously anticipated. Although there is little experimental evidence that ornaments influence hormone titers, theory suggests ornaments could affect the hormone titers of individuals displaying ornaments (signalers) and/or individuals perceiving ornaments (receivers) directly or indirectly via behavior.Understanding how ornaments influence hormone titers of signalers and receivers is important because hormones have persistent effects on morphology, physiology, and behavior. Androgens in vertebrates and juvenile hormone (JH) in insects mediate the fecundity vs. lifespan tradeoff (6, 7). High titers of androgens and JH typically increase success in competitive contexts, but also decrease survival (8, 9). Therefore, if ornaments alter hormones, the hormonal changes will produce a cascade of sub...